- •Extensive evolution and applicability of Artificial Intelligence in medicine.
- •Personalization and high diagnostic and therapeutic precision.
- •Crucial requirements of multicenter recruitment of large datasets.
- •Increasing biomarkers variability, to establish the potential clinical value of radiomics.
- •Development of robust explainable AI models.
1.1 AI in oncology
1.2 AI approaches using oncological biomarkers and radiomics
1.3 Interpretability of radiomics
2. Radiomics classification
2.1 Feature based radiomics
- 1.Shape features []: provide quantitative description of geometric properties of the ROIs/VOIs, such as surface area, total volume, diameter, sphericity or surface-to-volume ratio.
- 2.First order statistics (histogram-based features): describe the fractional volume for the selected region of voxels and the distribution of the voxels’ intensity, for example minimum, maximum, mean, variance, skewness, or kurtosis.
- 3.Second order statistics (textural features): These features are extracted based on matrices derived from intensity relationships of neighboring voxels in a 3D image [], such as:
- a.Gray Level Co-occurrence Matrix (GLCM): describes the spatial distribution of gray level intensities within a 3D image [].
- b.Gray Level Run Length Matrix (GLRLM): is defined as the number of contiguous voxels that have the same gray level value and it characterizes the gray level run lengths of different gray level intensities in any direction [].
- c.Gray Level Size Zone Matrix (GLSZM): quantifies gray level zones, the number of connected voxels that share the same gray level intensity, in a 3D image [].
- d.Neighbouring Gray Tone Difference Matrix (NGTMD): quantifies the difference between a gray value and the average gray value of its neighbours within a distance δ [].
- e.Gray Level Dependence Matrix (GLDM): quantifies the number of connected voxels within a distance δ that are dependent on the center voxel [].
- 4.Higher order statistics features: These features are obtained by statistical methods after applying filters or mathematical transformations to the image, in order to highlight repeating patterns, edges, histogram-oriented gradients, or local binary patterns of the segmentation. These include fractal analysis, Minkowski functionals, wavelet and Fourier transformations, as well as Laplacian transformations of Gaussian-filtered images, which can extract areas with increasingly coarse texture patterns [].
2.2 Deep learning radiomics (DLR) features
3. Deep learning
3.1 Neural networks and multilayer perceptron (MLP)
3.2 Convolutional neural networks (CNN)
3.3 Applications of deep learning in medical imaging
|Classification||Image classification||Skin Cancer||A Esteva et al. 2017 |
|Image classification||Diabetic retinopathy||V Gulshan et al. 2016 |
|Object classification||Lung Nodules||W Shen et al. 2015 |
|Object classification||Skin Lesions||J Kawahara et al. 2016 |
|Detection||3D translation to 2D classification||Bone localization||D Yang et al. 2015 |
|3D translation to 2D classification||Heart/Aorta localization||B de Vos et al. 2016 |
|Pixel wise classification||Histopathology||G Litjens et al. 2016 |
|Pixel wise classification||Coronary calcium scoring||JM Wolterink et al. 2016 |
|DL using 3D CNN||MRI Brain metastasis||O Charron et al. 2018 |
|DL using 3D CNN||MRI Brain Metastasis||E Grovik et al. 2020 |
|Segmentation||U-Net Convolutional Network||Neuronal structures in electron microscopy||O Ronneberger et al. 2015 |
|3D U-Net Convolutional Network||Volumetric segmentation Xenopus kidney||O Cicek et al. 2016 |
|V-Net: Fully Convolutional Network||MRI prostate volumetric Segmentation||F Milletari et al. 2016 |
|Multi-scale 3D CNN connected with Conditional Random Field||MRI Brain lesions (injuries, tumors, ischemic stroke)||K Kamnitsas et al. 2017 |
|Supervised 3D supervoxel learning||Multimodal MRI Brain tumor||M Soltaninejad et al. |
|Fully CNN combined with Non-quantifiable Local Texture Feature||MRI Brain tumor||W Deng et al. 2019 |
|Adaptive Neuro Fuzzy Inference System with Textural Features||Glioma Brain tumor||A Selvapandian et al. 2018 |
|Registration||CNN to derive transformation parameters||Neonatal brain tumor||M Simonovsky 2016 |
|CNN regression: Pose Estimation via Hierarchical Learning||Total Knee Arthroplasty Kinematics & X-ray transeophageal echocardiography||S Miao et al. 2016 |
|CNNs using artificial examples to adjust the transformation parameters||Lung radiotherapy||M Foote et al. 2019 |
|3D CNN||Proton Therapy prostate cancer||M Elmahdy et al. 2019 |
|Paul et al. |
|Lung nodules classification (malignant vs. benign)||Low dose CT||Three strategies were compared and combined: standard radiomics, pre-trained CNN and CNN trained from scratch with data augmentation.||Combining all three strategies led to the best performance|
|Ning et al. |
|gastrointestinal stromal tumors classification (malignant vs. benign)||CT||Standard radiomics vs. Pre-trained CNN based features, and combination into random forest||Combining both outperforms each approach separately|
|Antropova et al. |
|Breast lesions classification (malignant vs. Benign)||FFDM, US, DCE-MRI||Standard radiomics vs. Pre-trained CNN based features, and combination into support vector machine||Combination always led to best results in all the three image modalities|
|Diamant et al. |
|Head and neck cancer outcome prediction||CT||CNN trained from scratch on a 2D pre-segmented slice of the tumor (use of data augmentation by a factor of 20)||Slightly better performance using CNN compared to standard radiomics but not for all endpoints|
|Ypsilantis et al. |
|Esophageal cancer response to therapy prediction||PET||CNN trained from scratch on a set of fused 2D pre-segmented slices of the tumor (use of data augmentation by a factor of greater than 55)||Slightly better performance using CNN compared to standard radiomics|
|Hosny et al. |
|Lung cancer survival prediction||CT||3D CNN trained on pre-segmented volumes (use of data augmentation by a factor of 32000)||Slightly better performance of CNN over engineered features but not for all datasets.|
4. Explainable artificial intelligence (XAI)
4.1 Proxy models and model compression
4.2 Visualization of intermediate features
4.3 Importance estimators and relevance scores
Brocki L, Chung NC. Input Bias in Rectified Gradients and Modified Saliency Maps. 2021 IEEE International Conference on Big Data and Smart Computing (BigComp). https://doi.org/10.1109/BigComp51126.2021.00036.
5. Imaging data processing
5.1 Data curation
5.3 Harmonization in the image domain
5.3.1 Standardization of imaging procedures
5.3.2 Processing images
5.4 Harmonization in the feature domain
5.4.1 Selection of features based on their reliability
- Desseroit M.C.
- Tixier F.
- Weber W.A.
- Siegel B.A.
- Cheze Le Rest C.
- Visvikis D.
- et al.
5.4.2 Modifying the feature’s definitions
5.4.4 Batch effect removal
6. Discussion and conclusions
Declaration of Competing Interest
- Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries.CA Cancer J Clin. 2018; 68: 394-424
- Radiomics in radiation oncology-basics, methods, and limitations.Strahlenther Onkol. 2020; 196: 848-855
- Impact of whole-body 18F-FDG PET on diagnostic and therapeutic management of Medical Oncology patients.Eur J Cancer. 2008; 44: 1678-1683
- How Advances in Imaging Will Affect Precision Radiation Oncology.Int J Radiat Oncol Biol Phys. 2018; 101: 292-298
- Radiomics: the bridge between medical imaging and personalized medicine.Nat Rev Clin Oncol. 2017; 14: 749-762
- Radiomics: extracting more information from medical images using advanced feature analysis.Eur J Cancer. 2012; 48: 441-446
- there is more than meets the eye in medical imaging.Proceedings of the SPIE. 2016; 9785
- Radiomics: data are also images.J Nucl Med. 2019; 60: 38S-44S
- A survey on deep learning in medical image analysis.Med Image Anal. 2017; 42: 60-88
- A deep feature fusion methodology for breast cancer diagnosis demonstrated on three imaging modality datasets.Med Phys. 2017; 44: 5162-5171
- Deep Learning and Radiomics predict complete response after neo-adjuvant chemoradiation for locally advanced rectal cancer.Sci Rep. 2018; 8: 12611
- Deep learning in head & neck cancer outcome prediction.Sci Rep. 2019; 9: 2764
- ImageNet large scale visual recognition challenge.Int J Comput Vis. 2015; 115: 211-252
- A review of original articles published in the emerging field of radiomics.Eur J Radiol. 2020; 127108991
- Use of radiomics in the radiation oncology setting: Where do we stand and what do we need?.Cancer/Radiotherapie. 2020; 24: 755-761
- The Image Biomarker Standardization Initiative: Standardized Quantitative Radiomics for High-Throughput Image-based Phenotyping.Radiology. 2020; 295: 328-338
- Combined FET PET/MRI radiomics differentiates radiation injury from recurrent brain metastasis.Neuroimage Clin. 2018; 20: 537-542
- PET/MRI Radiomics inpatients with brain metastases.Front Neurol. 2020; 11: 1
- Radiomics: the process and the challenges.Magn Reson Imaging. 2012; 30: 1234-1248
- Radiomics: a new application from established techniques.Expert Rev Precis Med Drug Dev. 2016; 1: 207-226
- Different approaches for extracting information from the co-occurrence matrix.PLoS ONE. 2013; 8e83554
- Texture information in run-length matrices.IEEE Trans Im Proc. 1998; 7: 1602-1609
- Texture indexes and gray level size zone matrix. Application to cell nuclei classification.Pattern Recognition and Information Processing. 2009; : 140-145
- textural features corresponding to textural properties.IEEE Systems, Man and Cybernetics. 1989; 19: 1264-1274
- Neighboring gray level dependence matrix for texture classification.Comput Graph Image Process. 1982; 20: 297
- Radiomics: the facts and the challenges of image analysis.Eur Radiol Exp. 2018; 2: 36
- Deep Learning based Radiomics (DLR) and its usage in noninvasive IDH1 prediction for low grade glioma.Sci Rep. 2017; 7: 5467
- The role of deep learning and radiomic feature extraction in cancer-specific predictive modelling: a review.Transl Cancer Res. 2018; 7: 803-816
- Machine and deep learning methods for radiomics.Med Phys. 2020; 47: e185-e202
Kaiming H, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. arXiv:1512.03385 [cs.CV].
Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. In ‘Advances in Neural Information Processing Systems’, edited by F. Pereira and C. J. C. Burges and L. Bottou and K. Q. Weinberger 2012;25.
- Deep learning.Nature. 2015; 521: 436-444
- Learning representations by back-propagating errors.Nature. 1986; 323: 533-536
- A Stochastic Approximation Method.Ann Math Statist. 1951; 22: 400-407
- Reconciling modern machine-learning practice and the classical bias-variance trade-off.Proc Natl Acad Sci U S A. 2019; 116: 15849-15854
- A survey on image data augmentation for deep learning.J Big Data. 2019; 6: 60
- Dropout: a simple way to prevent neural networks from overfitting.J Machine Learning Res. 2014; 15: 1929-1958
Krogh A, Hertz JA. A Simple Weight Decay Can Improve Generalization. 4th International Conference on Neural Information Processing Systems: Morgan-Kaufmann. p. 950-7.
- Cross-Validatory Choice and Assessment of Statistical Predictions.J Royal Statistical Society: Series B. 1974; 36: 111-133
- Gradient-based learning applied to document recognition.Proc of the IEEE. 1998;
Le Cun Y, Boser B, Denker JS, Henderson D, Howard RE, Hubbard W, et al. Handwritten digit recognition with a back-propagation network. NeurIPS Proceedings. p. 396-404.
Scherer D, Muller A, Behnke S. Evaluation of pooling operations in convolutional architectures for object recognition. 20th International Conference on Artificial Neural Networks. Thessaloniki, Greece2010.
Springenberg JT, Dosovitskiy A, Brox T, Riedmiller M. Striving for simplicity: The all convolutional net. 2014.
- Deep convolutional neural networks for image classification: A comprehensive review.Neural Comput. 2017; 29: 2352-2449
Girshick R, Donahue J, Darrell T, J. M. Rich feature hierarchies for accurate object detection and semantic segmentation. 2014 IEEE Conference on Computer Vision and Pattern Recognition: IEEE; 2014. p. 580-7.
- You only look once: unified, real-time object detection.in: 2016 IEEE Conference on Computer Vision and Pattern Recognition: IEEE. 2016: 779-788
Noh H, Hong S, Han B. Learning deconvolution network for semantic segmentation. arXiv:1505.04366 [cs.CV]. 2015.
Chen LC, Papandreou G, Schroff F, Adam H. Rethinking atrous convolution for semantic image segmentation. arXiv:1706.05587 [cs.CV]. 2017.
- Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups.IEEE Signal Process Mag. 2012; 29: 82-97
Collobert R, Weston J. A unified architecture for natural language processing: deep neural networks with multitask learning. ICML '08: Proceedings of the 25th international conference on Machine learning2008. p. 160-7.
- Deep Learning in Medical Image Analysis.Annu Rev Biomed Eng. 2017; 19: 221-248
Liu X, faes L, Kale AU, Wagner SK, Fu DJ, Bruynseels A. A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis. The Lancet Digital Health. 2019;1:E271-E97.
- Dermatologist-level classification of skin cancer with deep neural networks.Nature. 2017; 542: 115-118
- Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs.JAMA. 2016; 316: 2402-2410
- Multi-scale Convolutional Neural Networks for Lung Nodule Classification.Inf Process Med Imaging. 2015; 24: 588-599
Kawahara J, Hamarneh G. Multi-resolution-Tract CNN with Hybrid Pretrained and Skin-Lesion Trained Layers. International Workshop on Machine Learning in Medical Imaging2016. p. 164-71.
Yang D, Zhang S, yan Z, Tan C, Li K, Metaxas D. Automated anatomical landmark detection ondistal femur surface using convolutional neural network. 2015 IEEE 12th International Symposium on Biomedical Imaging (ISBI); 2015. p. 17-21.
de Vos BD, Wolterink JM, Viergever MA, Isgum I. 2D image classification for 3D anatomy localization: employing deep convolutional neural networks. SPIE Medical Imaging; 2016.
Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation. arXiv:14114038 [csCV]; 2015.
- Deep learning as a tool for increased accuracy and efficiency of histopathological diagnosis.Sci Rep. 2016; 6: 26286
- Automatic coronary artery calcium scoring in cardiac CT angiography using paired convolutional neural networks.Med Image Anal. 2016; 34: 123-136
- Automatic detection and segmentation of brain metastases on multimodal MR images with a deep convolutional neural network.Comput Biol Med. 2018; 95: 43-54
- Deep learning enables automatic detection and segmentation of brain metastases on multisequence MRI.J Magn Reson Imaging. 2020; 51: 175-182
Ronneberger O, DFischer P, Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation. MICCAI 2015: Medical Image Computing and Computer-Assisted Intervention; 2015. p. 234-41.
Cicek O, Abdulkadir A, Lienkamp SS, Brox T, Ronneberger O. 3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation. MICCAI 2016 : Medical Image Computing and Computer-Assisted Intervention 2016.
Milletari F, Navab N, Ahmadi SA. V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation. arXiv:160604797v1 [csCV]; 2016.
- Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation.Med Image Anal. 2017; 36: 61-78
- Supervised learning based multimodal MRI brain tumour segmentation using texture features from supervoxels.Comput Methods Programs Biomed. 2018; 157: 69-84
- Brain Tumor Segmentation Based on Improved Convolutional Neural Network in Combination with Non-quantifiable Local Texture Feature.J Med Syst. 2019; 43: 152
- Fusion based Glioma brain tumor detection and segmentation using ANFIS classification.Comput Methods Programs Biomed. 2018; 166: 33-38
Simonovsky M, Gutierrez-Becker B, Mateus D, Navab N, Komodakis N. A Deep Metric for Multimodal Registration. MICCAI 2016: Medical Image Computing and Computer-Assisted Intervention2016. p. 10-8.
- A CNN Regression Approach for Real-Time 2D/3D Registration.IEEE Trans Med Imaging. 2016; 35: 1352-1363
Foote MD, Zimmerman BE, Sawant A, Joshi SC. Real-Time 2D-3D Deformable Registration with Deep Learning and Application to Lung Radiotherapy Targeting. Information Processing in Medical Imaging IPMI 20192019. p. 265-76.
- Robust contour propagation using deep learning and image registration for online adaptive proton therapy of prostate cancer.Med Phys. 2019; 46: 3329-3343
- Medical image registration using deep neural networks: A comprehensive review.Comput Electr Eng. 2020; 87106767
- Data are also images.J Nucl Med. 2019; 60: 38S-44S
- Machine (Deep) Learning Methods for Image Processing and Radiomics.IEEE Trans Rad Plasma Med Sci. 2019; 3: 104-108
- 3-D RPET-NET: Development of a 3-D PET Imaging Convolutional Neural Network for Radiomics Analysis and Outcome Prediction.IEEE Trans Radiation Plasma Med Sci. 2019; 3: 225-231
- Deep learning for lung cancer prognostication: A retrospective multi-cohort radiomics study.PLoS Med. 2018; 15e1002711
- Predicting Response to Neoadjuvant Chemotherapy with PET Imaging Using Convolutional Neural Networks.PLoS ONE. 2015; 10e0137036
- A Deep Learning-Based Radiomics Model for Prediction of Survival in Glioblastoma Multiforme.Sci Rep. 2017; 7: 10353
- Deep learning radiomics can predict axillary lymph node status in early-stage breast cancer.Nat Commun. 2020; 11: 1236
- Fully Automated Hybrid Approach to Predict the IDH Mutation Status of Gliomas via Deep Learning and Radiomics.Neuro Oncol. 2020;
- Pattern Classification for Gastrointestinal Stromal Tumors by Integration of Radiomics and Deep Convolutional Features.IEEE J Biomed Health Inform. 2019; 23: 1181-1191
- Predicting malignant nodules by fusing deep features with classical radiomics features.J Med Imaging (Bellingham). 2018; 5011021
Simonyan K, Zisserman A. Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv:14091556v6 [csCV]2015.
Doshi-Velez F, Kim B. Towards A Rigorous Science of Interpretable Machine Learning. arXiv:170208608v2 [statML]2017.
Gilpin LH, Bau D, Yuan BZ, Bajwa A, Specter M, Kagal L. Explaining Explanations: An Overview of Interpretability of Machine Learning. 2018 IEEE 5th International Conference on Data Science and Advanced Analytics (DSAA)2018. p. 80-9.
Hinton G, Vinyals O, Dean J. Distilling the Knowledge in a Neural Network. 2015.
- “Why Should I Trust You?”: Explaining the Predictions of Any Classifier. KDD '16.in: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2016: 1135-1144
Bucila C, Caruana R, Niculescu-Mizil A. Model ompression. KDD '06: Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining. 2006: 535–541.
Ba J, Caruana R. Do Deep Nets Really eed to be Deep? in Advances in Neural Information Processing Systems, edited by Z. Ghahramani and M. Welling and C. Cortes and N. Lawrence and K. Q. Weinberger. 2014:27.
Frosst N, Hinton G. Distilling a Neural Network Into a Soft Decision Tree. arXiv:1711.09784 [cs.LG]. 2018.
Distill. Feature Visualization: How neural networks build up their understanding of images. https://distill.pub/2017/feature-visualization/. Accessed on November 28, 2020.
Zeiler MD, Taylor GW, Fergus R. Adaptive Deconvolutional Networks for Mid and High Level Feature Learning. ICCV 20112011.
- Visualizing and Understanding Convolutional Networks.ECCV 2014: Computer Vision. 2014; : 818-833
- Evaluating the Visualization of What a Deep Neural Network Has Learned.EEE Trans Neural Networks Learning Syst. 2017; : 2660-2673
Fong RC, Vedaldi A. Interpretable Explanations of Black Boxes by Meaningful Perturbation. 2017 IEEE International Conference on Computer Vision (ICCV)2017. p. 3449-57.
Zintgraf LM, Cohen TS, Adel T, Welling M. Visualizing deep neural network decisions: prediction difference analysis. ICLR 20172017.
- How to Explain Individual Classification Decisions.J Machine Learning Res. 2010; 11: 1803-1831
Erhan D, Bengio Y, Courville A, Vincent P. Technical Report 1341: Visualizing Higher-Layer Features of a Deep Network. Universite de Montreal; 2009.
Smilkov D, Thorat N, Kim B, Viegas F, Wattenberg M. SmoothGrad: removing noise by adding noise. arXiv:170603825v1 [csLG]2017.
Kim B, Seo J, Jeon S, Koo J, Choe J, Jeon T. Why are Saliency Maps Noisy? Cause of and Solution to Noisy Saliency Maps. arXiv:190204893 [csLG]2019.
Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization. 2017 IEEE International Conference on Computer Vision (ICCV)2017. p. 618-26.
Brocki L, Chung NC. Input Bias in Rectified Gradients and Modified Saliency Maps. 2021 IEEE International Conference on Big Data and Smart Computing (BigComp). https://doi.org/10.1109/BigComp51126.2021.00036.
Abebayo J, Gilmer J, Muelly M, Goodfellow I, Hardt M, Kim B. Sanity Checks for Saliency Maps. arXiv: 1810.03292 [cs.CV]. 2018.
- Evaluating the Visualization of What a Deep Neural Network Has Learned.IEEE Trans Neural Netw Learn Syst. 2017; 28: 2660-2673
Hooker S, Erhan D, Kindermans PJ, Kim B. A Benchmark for Interpretability Methods in Deep Neural Networks. arXiv:180610758v3 [csLG].
- Preparing Medical Imaging Data for Machine Learning.Radiology. 2020; 295: 4-15
- Data Curation for Preclinical and Clinical Multimodal Imaging Studies.Mol Imaging Biol. 2019; 21: 1034-1043
- Free DICOM de-identification tools in clinical research: functioning and safety of patient privacy.Eur Radiol. 2015; 25: 3685-3695
van Herk M. Conquest DICOM software. http://ingenium.home.xs4all.nl/dicom.html. Accessed on January 26, 2021.
RSNA. CTP-The RSNA Clinical Trial Processor. http://mircwiki.rsna.org/index.php?title=CTP-The_RSNA_Clinical_Trial_Processor. Accessed on January 26, 2021.
Knopke A. K-Pacs. http://k-pacs.net/. Accessed on January 26, 2021.
Library D. DICOM Library - Anonymize, Share, View DICOM files ONLINE. http://www.dicomlibrary.com/. Accessed on January 26, 2021.
DicomWorks. DicomWorks - Free DICOM software. http://www.dicomworks.com/. Accessed on January 26, 2021.
Publishing P. PixelMed Java DICOM Toolkit. http://www.pixelmed.com/. Accessed on January 26, 2021.
DVTk. DVTk Project. http://www.dvtk.org/. Accessed on January 26, 2021.
Yakami M. YAKAMI DICOM Tools. http://www.kuhp.kyoto-u.ac.jp/%7Ediag_rad/intro/tech/dicom_tools.html. Accessed on January 26, 2021.
NIfTI. Neuroimaging Informatics Technology Initiative. https://nifti.nimh.nih.gov/. Accessed on January 26, 2021.
- Temporal stability assessment in shear wave elasticity images validated by deep learning neural network for chronic liver disease fibrosis stage assessment.Med Phys. 2019; 46: 2298-2309
- Deep learning networks on chronic liver disease assessment with fine-tuning of shear wave elastography image sequences.Phys Med Biol. 2020; 65215027
- A decade of radiomics research: are images really data or just patterns in the noise?.Eur Radiol. 2020;
- Applications and limitations of radiomics.Phys Med Biol. 2016; 61: R150-R166
- Radiomics in nuclear medicine: robustness, reproducibility, standardization, and how to avoid data analysis traps and replication crisis.Eur J Nucl Med Mol Imaging. 2019; 46: 2638-2655
- Why validation of prognostic models matters?.Radiother Oncol. 2018; 127: 370-373
- Characterization of PET/CT images using texture analysis: the past, the present... any future?.Eur J Nucl Med Mol Imaging. 2017; 44: 151-165
- Imaging biomarker roadmap for cancer studies.Nat Rev Clin Oncol. 2017; 14: 169-186
- Distributed learning: Developing a predictive model based on data from multiple hospitals without data leaving the hospital - A real life proof of concept.Radiother Oncol. 2016; 121: 459-467
- Variability of textural features in FDG PET images due to different acquisition modes and reconstruction parameters.Acta Oncol. 2010; 49: 1012-1016
- Prediction of cervical cancer recurrence using textural features extracted from 18F-FDG PET images acquired with different scanners.Oncotarget. 2017; 8: 43169-43179
- Impact of Image Reconstruction Settings on Texture Features in 18F-FDG PET.J Nucl Med. 2015; 56: 1667-1673
- A Postreconstruction Harmonization Method for Multicenter Radiomic Studies in PET.J Nucl Med. 2018; 59: 1321-1328
- EANM/EARL harmonization strategies in PET quantification: from daily practice to multicentre oncological studies.Eur J Nucl Med Mol Imaging. 2017; 44: 17-31
- FDG PET/CT: EANM procedure guidelines for tumour imaging: version 2.0.Eur J Nucl Med Mol Imaging. 2015; 42: 328-354
- Feasibility of state of the art PET/CT systems performance harmonisation.Eur J Nucl Med Mol Imaging. 2018; 45: 1344-1361
- Experimental Multicenter and Multivendor Evaluation of the Performance of PET Radiomic Features Using 3-Dimensionally Printed Phantom Inserts.J Nucl Med. 2020; 61: 469-476
- A decade of radiomics research: are images really data or just patterns in the noise?.Eur Radiol. 2021; 31: 1-4
- Can alternative PET reconstruction schemes improve the prognostic value of radiomic features in non-small cell lung cancer?.Methods. 2020;
Roboflow. You might be resizing your images incorrectly. https://blog.roboflow.com/you-might-be-resizing-your-images-incorrectly/. Accessed on November 28, 2020.
- Validation of a harmonization method to correct for SUV and radiomic features variability in multi-center studies.J Nucl Med. 2018; 59: 288
ADNI. Pet Acquisition. http://adni.loni.usc.edu/methods/pet-analysis-method/pet-analysis/. Accessed on November 28, 2020.
- Deep Learning-based Image Conversion of CT Reconstruction Kernels Improves Radiomics Reproducibility for Pulmonary Nodules or Masses.Radiology. 2019; 292: 365-373
Hognon C, Tixier F, Gallinato O, Colin T, Visvikis D, Jaouen V. Standardization of Multicentric Image Datasets with Generative Adversarial Network. IEEE MIC2019.
Modanwal G, Vellal A, Buda M, Mazurowski MA. MRI image harmonization using cycle-consistent generative adversarial network. SPIE Medical Imaging 20202020.
- Inter-site harmonization based on dual generative adversarial networks for diffusion tensor imaging: application to neonatal white matter development.Biomed Eng Online. 2020; 19: 4
- Normalization of multicenter CT radiomics by a generative adversarial network method.Phys Med Biol. 2020;
- Reliability of PET/CT Shape and Heterogeneity Features in Functional and Morphologic Components of Non-Small Cell Lung Cancer Tumors: A Repeatability Analysis in a Prospective Multicenter Cohort.J Nucl Med. 2017; 58: 406-411
- Development of a nomogram combining clinical staging with (18)F-FDG PET/CT image features in non-small-cell lung cancer stage I-III.Eur J Nucl Med Mol Imaging. 2016; 43: 1477-1485
- Impact of image preprocessing on the volume dependence and prognostic potential of radiomics features in non-small cell lung cancer.Transl Cancer Res. 2016; 5: 349-363
- Intrinsic dependencies of CT radiomic features on voxel size and number of gray levels.Med Phys. 2017; 44: 1050-1062
- Voxel size and gray level normalization of CT radiomic features in lung cancer.Sci Rep. 2018; 8: 10545
- Harmonization strategies for multicenter radiomics investigations.Phys Med Biol. 2020;
- Learning from scanners: Bias reduction and feature correction in radiomics.Clin Transl Radiat Oncol. 2019; 19: 33-38
- Neural network training for cross-protocol radiomic feature standardization in computed tomography.J Med Imaging (Bellingham). 2019; 6024008
- Creating robust predictive radiomic models for data from independent institutions using normalization. IEEE Trans Radiat Plasma.Med Sci. 2019; 1
- Radiomic analysis for pretreatment prediction of response to neoadjuvant chemotherapy in locally advanced cervical cancer: A multicentre study.EBioMedicine. 2019; 46: 160-169
- Adjusting batch effects in microarray expression data using empirical Bayes methods.Biostatistics. 2007; 8: 118-127
- Why Batch Effects Matter in Omics Data, and How to Avoid Them.Trends Biotechnol. 2017; 35: 498-507
- Minimizing acquisition-related radiomics variability by image resampling and batch effect correction to allow for large-scale data analysis.Eur Radiol. 2020;
- Validation of A Method to Compensate Multicenter Effects Affecting CT Radiomics.Radiology. 2019; 291: 53-59
Orlhac F, Lecler A, Savatovski J, Goya-Outi J, Nioche C, Charbonneau F, et al. How can we combat multicenter variability in MR radiomics? Validation of a correction procedure. Eur Radiol. 2020.
- External validation of a combined PET and MRI radiomics model for prediction of recurrence in cervical cancer patients treated with chemoradiotherapy.Eur J Nucl Med Mol Imaging. 2019; 46: 864-877
- Pretreatment (18)F-FDG PET/CT Radiomics Predict Local Recurrence in Patients Treated with Stereotactic Body Radiotherapy for Early-Stage Non-Small Cell Lung Cancer: A Multicentric Study.J Nucl Med. 2020; 61: 814-820
- Harmonization of radiomic features of breast lesions across international DCE-MRI datasets.J Med Imaging (Bellingham). 2020; 7012707
- Radiomics Analysis of Computed Tomography helps predict poor prognostic outcome in COVID-19.Theranostics. 2020; 10: 7231-7244
- External validation of radiomics-based predictive models in low-dose CT screening for early lung cancer diagnosis.Med Phys. 2020;
- Performance comparison of modified ComBat for harmonization of radiomic features for multicenter studies.Sci Rep. 2020; 10: 10248
Parekh VS, Jacobs MA. Radiomic synthesis using deep convolutional neural networks. arXiv:1810.11090 [cs.CV]. 2018.
- Repeatability and Reproducibility of Radiomic Features: A Systematic Review.Int J Radiat Oncol Biol Phys. 2018; 102: 1143-1158
KERAS. Deep Learning with KERAS Radiomics. https://github.com/decordoba/deep-learning-with-Keras-Radiomics) Accessed on Jan. 29, 2021.
Abadi M, Barham P, Chen J, Chen U, Davis A, Dean J, et al. Tensorflow: a system for large-scale machine learning. 12th symposium on operating systems design and implementation2016. p. 265-83.
- LIFEx: A Freeware for Radiomic Feature Calculation in Multimodality Imaging to Accelerate Advances in the Characterization of Tumor Heterogeneity.Cancer Res. 2018; 78: 4786-4789
- MaZda–a software package for image texture analysis.Comput Methods Programs Biomed. 2009; 94: 66-76
- Computational Radiomics System to Decode the Radiographic Phenotype.Cancer Res. 2017; 77: e104-e107
User LicenseCreative Commons Attribution (CC BY 4.0) |
- Read, print & download
- Redistribute or republish the final article
- Text & data mine
- Translate the article
- Reuse portions or extracts from the article in other works
- Sell or re-use for commercial purposes
Elsevier's open access license policy